nagatoyuki0943 / anomalib-patchcore-libtorch Goto Github PK

您好，您在anomalib中使用方法.md中写的
It is possible to export your model to ONNX or Torchscript or OpenVINO IR

If you want to export your PyTorch model to an OpenVINO model, ensure that export_mode is set to "openvino" in the respective model config.yaml.

导出openvino会先导出onnx, torchscript再导出openvino

导出onnx会一同导出torchscirpt-

我用fastflow测试，无论export_mode写openvino还是onnx，都无法导出torchscript,请问大概是哪儿的问题，拜谢！

auto result = inference(model, x);推理的时候,这句一直报错.
vs显示下面这个
0x00007FFE05023B19 处(位于 TorchProject1.exe 中)引发的异常: Microsoft C++ 异常: c10::Error，位于内存位置 0x0000009E6F3FB080 处。
Debug Error!
Program: D:\Game\TorchProject1\x64\Debug\TorchProject1.exe abort() has been calle
程序显示在IListRef.h出现了断点,即下面这里
#if defined(_MSC_VER) && _ITERATOR_DEBUG_LEVEL == 2
// See [Note: MSVC Iterator Debug]
~IListRefIterator() {
switch (tag_) {
case IListRefTag::Boxed:
payload_.boxed_iterator.~boxed_iterator_type();
case IListRefTag::Unboxed:
payload_.unboxed_iterator.~unboxed_iterator_type();
default:
TORCH_INTERNAL_ASSERT(false, "invalid IListRef tag.");
}
您能帮我看看哪里有问题吗?